PASTURE TOP-DRESSING EXPERIMENTS IN OTAGO, SEASON 1927-28.

New Zealand Journal of Agriculture, Volume XXXVI, Issue 3, 20 March 1928, Page 173

PASTURE TOP-DRESSING EXPERIMENTS IN OTAGO, SEASON 1927-28.

R. B. Tennent,

.D.D., Instructor in Agriculture, and A. A. Hume, A.R.C.Sc.I

Assistant Instructor in Agriculture, Dunedin

During the past few years considerable interest has been evinced in Otago in regard to the practice of top-dressing pastures. The results claimed by the application of fertilizers in different parts of the Dominion naturally caused many Otago farmers to suppose that similar results could be obtained in their province, and as a consequence the Fields Division in Dunedin has been inundated with inquiries asking for specific information as to the advisability of adopting the practice of top-dressing, and' discarding the well-established practice of ploughing up old pasture and sowing down young grass. Sufficient general information had already been collated by the instructional staff of the province to indicate that in some localities remunerative returns could be obtained from the application of certain fertilizers, to pastures which had apparently outlived their usefulness, but on examination this information proved indefinite and in many •cases open to doubt. It was therefore decided to inaugurate an extensive series of experiments on top-dressing, these to cover a'wide area and to embrace within their scope a large number of ! soil-types and pastures of varying ages and condition. To this end the co-operation of the New Zealand Farmers’ Union was enlisted, and as a result forty-two co-operative experimental plots devoted solely to top-dressing experiments were laid down and cut for hay during the past season. The wide distribution of these plots can be seen from the accompanying map (next page).

In arranging the scope of the experiment it was decided to adhere to a uniform plan throughout, using the drill strip method, thus allowing all results to be examined statistically. The method of laying out the plots was practically the same as that used by Mr. A. W. Hudson for the Fields Division experimental work in Canterbury, which has been fully described in the Journal. With the exception of three plots which were laid down during the season 1926-27 all plots received similar treatment. The plan of the experiments was comparatively simple, being to test the efficacy of phosphates both alone and in conjunction .with lime, and further to acquire definite information as to the value of lime, alone on pastures.

The phosphates under trial were superphosphate (44-46. per cent, .grade) and basic slag (Trifolium brand, 17-20 per cent, grade). These manures were applied at a uniform rate of 3 cwt. per acre, and to -ensure accuracy in this respect a drill belonging to the Department was utilized throughout. Carbonate of lime in a finely ground form was used at the rate of 1 ton, per acre. The plots were dressed with the various treatments during the months of July, August, and September. Some of the dressings were therefore given at a comparatively late period, and this fact has to be taken into consideration in examining the first year’s results. At an appropriate period the plots were shut up for hay. On account of the large number of plots to be harvested it was naturally

not feasible to get each plot dealt with at its peak point of .production. This accounts for plots situated on high-class land in some cases showing comparatively low weighings when compared with those on poor soil-types. The yields given throughout in no way represent the relative fertility of the different plots.

The only gauge utilized in estimating the effect of the various fertilizers was the green-weight increase of the. treated plots over the untreated or “ control ” plots. This obviously is not an ideal method of estimating the effect of top-dressing, but when taken in conjunction with the character of the pasture constituents on the various plots it affords a reasonable indication if any one treatment or combination of treatments is giving better results than untreated plots.

An endeavour has been made to present the results in a manner intelligible to farmers, for whom this investigation has been primarily

conducted. To this end all green weights harvested have been converted into estimated hay weights, and the yield per acre of hay is thus shown. An arbitrary value of £5 per ton has been given to alb hay harvested, and from that has been deducted the value of the crop. Against this value the cost of the various fertilizers has been placed, and a profit or loss table over the control plot worked out.

In presenting these tables it is to be noted that the hay values only represent the amount of material harvested during the comparatively short period of time (about ten weeks on the average) during which the plots were closed to stock. The full cost of the fertilizers used has for convenience been debited against the hay thus produced. No residual effect has been taken into account, and it is only reasonable to expect that for the next few years the effect of certain fertilizers will be noticed on some portions of the plots. Again, it is to be pointed out that thirty-nine of the plots are in their first year of treatment, consequently no great effect from lime can as yet be expected. It is intended to carry on the experiment with these plots for a number of years in order to ascertain the length of time over which the effects of the different fertilizers used will be felt. For the next few years, therefore, 'the results of these experiments will be presented to farmers, and the profit or loss’ shown on each plot will be adjusted from year to year. For convenience this season’s results are divided into groups —namely, North Otago, Central Otago, and South Otago. North Otago. (l) J. G. WILLIAMSON, BORTON'S. ' This paddock, situated on rolling country adjacent to the Waitaki River, had been sown down with rape in 1923, no manure being used. The pasture before top-dressing had run largely to goose-grass, sweet vernal, and crested dogstail, ' small quantities of rye-grass, cocksfoot, and red and white clover showing through. The plot was top-dressed on 17th August, 1927, closed to stock on Ist October, and harvested sth December. Results are shown in the following table: —

Summary: The effect of superphosphate on this plot was very striking ; red clover responded to it most vigorously. No apparent results in the composition of the pasture could be noted at time of

cutting from either lime or basic slag. This plot was kept closed to allow the aftermath to come away, and again the strips receiving superphosphate showed up most vividly. On the aftermath more clover growth could be noticed in .the basic-slag strips than in the control strips ■ ' (2) R. G. BORRIE, PAPAKAIO. The pasture utilized in this experiment is situated at the foothills of Papakaio, on a heavy piece of country locally described as a “ tarry ” soil, typical of a small area in the immediate vicinity. Sown down in 1904 with a mixture of rye-grass and red and white clover, it still held a fair sole of rye-grass and clover. Crested dogstail, sweet vernal, and other, poorer types of vegetation, however, were gradually replacing the good grasses. No manure has been used on this pasture since it was sown down over twenty-three years ago. The plot was top-dressed on 23rd August, 1927, closed to stock Ist October, and. harvested 6th December. Results were as follows :

Summary: As will be noted, superphosphate with lime gave the greatest increase over control.' Lime alone showed a significant response, as also basic slag alone and basic slag with lime. It was observed that those strips top-dressed with super alone and super with lime showed a much heavier clover content than all other strips.

(3) a. C. HURST, WINDSOR. This plot represented a considerable area of typical low hill country of the Windsor district, being situated on quite good land. The pasture had been sown down with rape in 1922, a meat-works manure being used at the rate of 1 cwt. per acre. The grasses used were cocksfoot, perennial- rye-grass, crested dogstail, and cow-grass. The pasture showed distinct. signs of deterioration, inferior grasses appearing in the sward. The cow-grass had given way to white clover. The plot was top-dressed on 18th August, 1927, closed 12th October, and harvested 20th December. Table 3 (next page) shows the results. Summary: No great weight increases were observed on this plot, but there was an undoubted increase in white-clover content on those

strips receiving superphosphate. All differences in yields from the various fertilizers are significant, and, as will be observed, in this case super gave the highest weight increase. The effect of lime was practically negligible. (4) MRS. MANSON, ENFIELD. This plot is located on a typical “ tarry,” heavy soil at Enfield. The pasture had been laid down about 1917, the grasses used being chiefly cocksfoot, rye-grass, crested dogstail, and red clover. At the date of laying down the experimental plot the pasture had chiefly gone to cocksfoot, crested dogstail, and white clover, with a large proportion of inferior grasses and weeds showing through. The plot was topdressed on 19th August, 1927, closed Ist October, and harvested Bth December. Results are tabulated below :

' Summary: On harvesting this plot the super strips could be clearly defined on account of the increase of white clover thereon. Super and super with lime gave the heaviest weighings, forming quite a good sole on the strips so treated. No other treatments showed visible response, despite the fact that when their weights were analysed slight significant increases were recorded.

(□) w. MCMILLAN (WINDMILL PADDOCK), HERBERT.

This pasture had been sown down in 1925 with rye-grass, crested dogstail, and red and white clover. Although naturally decreasing in rye-grass content, it is still well covered with this grass. The pasture, therefore, when top-dressed was in good condition. The plot is situated on low country adjacent to the coast. Top-dressing took place on 24th August, 1927; the plot was closed Ist October, and harvested 9th January, 1928. Table 5 gives results.

Summary : A general increase of clover from all treatments was noted on this plot, a combination of, superphosphate with lime giving the heaviest yield. Lime alone gave quite a significant increase over the unlimed plots, and in this respect it is to be noted that the pasture is comparatively young. The results from this plot should afford some interesting points next season. (6) W. MCMILLAN, HERBERT. This pasture had been sown down in the autumn of 1921 with oats, the grasses used being rye, dogstail, and red and white clover; no manure was used. The pasture had deteriorated considerably, a fair proportion of Yorkshire fog. and' brown-top taking charge. The plot was top-dressed on 24th August, 1927, closed Ist October, and harvested 10th January, 1928. Results were as under: —

Summary : This plot was really not advanced enough for harvesting, being extremely slow in making recovery. ' The weights throughout were poor, and in no way indicate the possibilities of this pasture. A perusal of the results will show that although there is little difference in the yields recorded for the various treatments the differences are significant. This would indicate that under more favourable conditions of growth greater differences' between treatments would be recorded. The plot will be closed to stock at a more opportune time next season. It should be noted, however, that the six strips top-dressed with superphosphate stood out quite plainly, on account of the greater proportion of white clover growing upon- them. (7) W. S. TROTTER, HILLGROVE. Sown down about 1902, this pasture had run mainly to sweet vernal with a sprinkling of white clover throughout. The. pasture was miserably poor, and little result was anticipated from the manurial treatment. Top-dressing took place on 15th August, 1927; the plot was closed Ist October, and harvested 9th December. The following table gives results: —■

Summary : Excellent results were obtained from superphosphate, the recovery of white clover being most marked. It appeared incredible that a response of this degree could be obtained on such a run-out pasture, the superphosphate strips . with their dense mats of clover showing out most clearly. Slag also responded, but to a much less visible degree than super. Although an increase in weight was obtained as a result of liming, no visible difference could be noted. This plot was kept closed to allow an aftermath to grow, and again the strips top-dressed with super showed out most markedly.

(8) H. S. SHEAT, BUSHEY. The pasture upon which this plot was situated is reported to 'be over twenty-five years of age. For such an old pasture its condition prior to top-dressing was ■ very satisfactory, there ■ being a good sole of rye-grass and white clover. Naturally, a fair proportion of brown-top and crested dogstail showed throughout. The plot was top-dressed on

25th August, 1927, closed Bth October, and harvested Bth December. Results are presented below: —

Summary: Superphosphate in conjunction with lime gave the highest return, a much closer sward of white clover being observed in this treatment. No apparent differences could be noticed in the cross-dressings of lime ; ' an increase, however, was recorded in their weighings. ■ • As will be noted, all treatments gave a significant increase.

(g) J. A. PAYNE, WAIKOUAITI. The pasture selected for this trial was laid down in 1915. Since that date it has been used largely as a horse-paddock, much coarse, rank vegetation growing upon it. From a pasture viewpoint this paddock was in poor condition, on account of the prevalence of brown-top and rank fog and cocksfoot. White clover was interspersed throughout. Top - dressing took place on 25th August, 1927; the plot was closed Ist October, and harvested 19th December. Table 9 gives results.

Summary : At the date of harvesting the plot showed a very uneven appearance on account of the great variability of the soil-fertility. This can be largely attributed to the horse-manure scattered throughout, and emphasizes the difficulty of obtaining uniform results on a pasture of this nature. As will be observed from the table, small differences were recorded. A distinct difference in clover content could be observed in those strips top-dressed with superphosphate. More definite results can be looked for from this experiment next season, as it is Mr. Payne’s intention to keep horses off the plot. The removal of the coarse grass will also result in more even growth. (io) J. b. CHAPMAN, WAITAKI. This was an extremely interesting experiment on an old pasture dating from about 1892. The pasture, situated on the gravel terrace of the Waitaki River, had run to Danthonia pilosa, sweet vernal, goosegrass, and suckling-clover. Numerous other grasses and plants were prevalent throughout the pasture, all of an inferior nature. Definite information was sought in regard to the effect of top-dressing this type of grassland, and a plot was accordingly laid down on 22nd August, 1927, and closed on 20th October. The plot was kept under close observation for a period of twelve weeks. Very little growth took place on any of the strips. The only difference which could be observed was a distinct increase in sucklingclover on those strips top-dressed with superphosphate. The growth on the whole plot was too light to harvest consequently observations only could be made. No effect could be observed from basic slag or lime. On this year’s results even the increase in suckling-clover could

not justify top-dressing. The plot will again be closed for observation next season. In the meantime, however, it appears that top-dressing an old run-out pasture on this type of soil will not pay. These remarks do not apply to young grass, Mr. Chapman having obtained' encouraging results from his own experiments on new pastures. (ll) R. F. SANDERS, STEWART SETTLEMENT. This pasture is a purely native one, mainly composed of Danthonia pilosa, into which had encroached such species as goose-grass, English hair-grass, and sweet vernal, while a small amount of suckling-clover could be noticed throughout. The pasture is situated on the old gravel terrace of the Waitaki River, and the experiment was conducted to afford information as to the effect of top-dressing this class of grassland. In many respects the plot resembled that of Mr. J. B. Chapman (No. io), and the results obtained were practically similar. Topdressed on 16th August, 1927, the plot was closed to stock on Ist October. Close observation was made at various times throughout the period. A fair general growth took place over the plot, but proved too light to harvest. A definite increase in suckling-clover content on those strips top-dressed with superphosphate could be noted. No other treatment appeared to have taken effect. The increase of suckling-clover in this year’s results could not justify the expense of top-dressing. Bearing in mind the dryness of this class of country it will be interesting to observe next season’s results.

Number | Paired . Manure. Mean Green Weight of Plot. Significant (S) or Non-signifi-cant (N). Estimated Weight of Hay per Acre. Value of Hay per Acre. Cost of Manure per Acre. Profit or Loss compared with Unmanured Plot. - Significant (S) or Non-signifi-cant (N). Estimated Weight of Hay per Acre. Value Hay p Acre. of 31 Cost of Manure per Acre. , Profit or Loss compared with Unmanured Plot. lb. T. cwt. qr. £ s. d. d. £ s. d. £ s. d. £ s. d. £ s. d. 30 Basic slag 6-8 N O IO I 2 113 3 0 15 0 O 15 O 013 (loss) 013 (loss) 30 Basic slag and lime 7’9 N O 12 O 3 0 0 0 1 8 0 I 8 0 056 (loss) 0 5 6 (loss) 24 Superphosphate . . 2Q-I S 24O II O O 0 1 1 0 1 10 816 (gain) 816 (gain) 24 Super and lime . . 28-7 S 232 10 17 6 6 1 14 0 1 14 0 760 (gain) 760 (gain) 44 Lime 6-4 * S ■ 093 289 9 0 13 0 0 13 0 019 (loss) 019 (loss) Control. . 4-9 0 72 I 17 6 6

Table I.

of Plots. Paired Manure. Mean Green Weight of Plot. Significant (S) Non-signifi- or (N). Estimated Weight of Hay per Acre. Value of Hay per Acre. Cost of Manure per Acre. Profit or Loss compared with Unmanured Plot lb. T. cwt. qr. cwt. £ s. d. qr. £ s. d. £ £ x s. d. s. d. £ s. d. £ s. d. 30 Basic slag and lime 18-4 S O 16 O 16 4 O O' 0 1'8 O' 4 0 10 6 (loss) 0 0 I 8 0 0 IO 6 (loss) 30 Basic slag 16-6 S O 14 2 14 3 12 6 2 0 15 0 3 0 5 0 (loss) 12 6 0 15 0 0 5 0 (loss) 30 Superphosphate . . 29-4 S I 6'0 6 6 10 0 0 I I ; O 6 2 6 6 (gain) IO 0 I 1 0 2 6 6 (gain) 3° Super and lime . . 34’7 . • S I IO I IO 7 1* 3 I I 14 O 7 2 14 9 (gain) 11 3 I 14 0 2 14 9 (gain. 40 Lime ... 17-9 S 0 15 2 15 3 17 6 2 O 13 O 3 020 (gain 17 6 0 13 0 0 2 0 (gain) Control. . 14-4 O 12 2 12 326 2 3 * * 2 6

Table 2.

Number Paired Manure. I Mean j Weight!Green of Plot. (S) Non-signifi- or cant Estimated Weight of Hay per Acre. Value of Hay per Acre. Cost of Manure per Acre. Profit or Loss compared with Unmanured Plot. 1 ; Profit or Loss ■ compared with Unmanured Plot. • / lb. T. cwt. qr. £ s. d. £ s. d. £ s. d. 36 ■ Basic slag . . ■ 27-4 s s 140 600 015 O 026 (loss) 30 Basic slag and lime 28-7 s i 5 0 650 180 0 10 6 (loss) 36 .Superphosphate . . 34-° s 19 3. 789 I I 0 1 0 3 (gain) 30 Super and lime . . I- 9 s f 7 3 6189 1 14 0 029 (loss) 32 Lime 26-3 s 1 3 0 5 1.5 0 013 O 056 (loss) Control. . 26-3 24’7 S. 1 3 0 1 1 2 5 1.5 0 5 7 6 0 13 0 056 (loss)

Table 3.

Number of Plots. Manure. Mean Weight Plot. (S) or (N). Estimated Weight of Hay per Acre. Value of Hay per Acre. Cost of Manure per Acre. Profit or Loss compared with Unmanured Plot. lb. 1’ T. cwt. qr. T. cwt. qr. £ s. d. £ ' s. d. £ s - d. £ s. ' d. £ s. d. £ s. d. 30 Basic slag 19 ’5 S 017 O. 4 5 0 O 15 O 050 (loss) 30 Basic slag and lime 197 s O 17 I 463 180 016 9 (loss) 30 . Superphosphate . . 36-9 s I 12 I 8 1 3 I I O 3 5 3 (gain) 30 Super and lime . . 35-o s I. IO 2 712 6 I 14 O 236 (gam) 4° Lime 17-6 s O 15 1 3 16 3 0 13 0 on 9 (loss) Control. . 17-3 O 1.5 0 3 15 0

Table 4

Number Plots. Manure. Mean Green. Weight of tn cd Estimated Weight (S) of Hay or Non-signifi-cant (N). Estimated Weight of Hay per Acre. Value of Hay per Acre. Cost of Manure per Acre. Profit or Loss compared with Unmanured Plot. d rt ’c b c7 cU V) '— g d 0 ' lb. T. cwt. qr. £ s. d. £ s. d. £ s. d. 30 Basic slag 24-9 S i i 3 589 0 15 0 039 (loss) 30 Basic slag and lime 28-8 S 1 5 1 663 180 009 (gain) 30 Superphosphate . . 28-6 S 1 5 0 650 I I O 066 (gain) 30 Super and lime . . 29-4 S 1 5 3 689 1140 029 (loss) 40 Lime 26-8 S’ 1 3 2 5 17 6 O 13 O 070 (gain) Control. . ■ . . 22-4 019 4 17 6 • * ’

Table 5

of Plots. Manure. Mean Green Weight of Estimated Weight (S) of Hay or Non-signifi-cant (N). Estimated Weight of Hay per Acre. Value of Hay per Acre. Cost of Manure per Acre. Profit or Loss compared with Unmanured Plot 1 c Ct C £ *c b a or Non-signi cant (N). ' - lb. T. cwt qr. qr. £ s. d. £ £ s. d. C £ s. d. d. £ s. d. £ s d. 30 Basic slag I 4 -I S on 0 0 2 15 0 2 O 15 O 15 0 17 6 (loss) 0 0 15 0 0 17 6 (loss) 30 Basic slag and lime 15-5 S 0 12 0 0 3°o 3 180 0 156 (loss) 0 I 8 0 I 5 6 (loss) 30 Superphosphate . . 16-o S 0 12 2 2 3'2 6 3 I 1.0 2 0 16 0 (loss) 6 I 1 0 0 16 0 (loss) 30 Super and lime . . 16-5 S 0 13 0 0 3 5 0 3 1 14 0 5 16 6 (loss) 0 I 14 0 I 6 6 (loss) 40 Lime L5-7 S O 12 O 12 I 3 1 3 I 0 13 0 3 093 (loss) 0 13 0 0 9 3 (loss) Control. . •• M-5 Control. . . . I 14-5 3 II on 2 2 17 6 17 6

Table 6

ofNumber [' Plots. Manure. Mean Green Weight Plot. (S) Non-signifi- or (N). cant Estimated Weight of Hay per Acre. Value of Hay per Acre. ■ Cost of Manure per Acre. Profit or Lose compared with Unmanured Plot ■ lb. T cwt. qr. cwt £ s. d. qr.’ £ s. d. £ £ s. d. s. d. £ s. d. £ s. d. 30 Basic slag 21-4 S O 18 3 18 4 13 9 3 O 15 O 4 050 (loss) 13 9 0 15 0 0 5 0 (loss) 30 Basic slag and lime 25-0 S I i 3 . I 5 8 9 3 180 5 030 (loss) 8 9 1 8 0 0 ' 3 0 (loss) 30 ' Superphosphate . . 34'i s I 9 3 9 789 3 I I O 7 2 4 0 (gain) 8 9 1 1 0 2 4 0 (gain) 30 Super and lime . . 35'2 s I 10 3 IO 7 13 9 3 1140 7 1 16 0 (gain) 13 9 1 14 0 I 16 0 (gain) 44 ■ Lime 21-8 S 019 I 19 4 16 3 1 O 13 O 4 006 (loss) 16 3 • 0 13 0 0 0 6 (loss) Control. . 19-3 0 16 3 > 6 , 4 3 9 3 4 3 9

Table 7.

of Numter 1 Plots. Manure. Mean Green of Plot. Significant 'Nqn-signifi- or (N). cant Estimated Weight of Hay ■ per Acre. Value of Hay per Acre. Cost of Manure per Acre. . Profit or Loss compared with Unmanured Plot. lb. T. cwt. qr. £ s. d. £ s. d. £ s. d. ■ 30 Basic slag 37’7 S " I 13 0 850 O 15 0 . 039 (gain) 30 Basic slag and lime 39-2 S I 14 I 8113 180 030 (loss) 30 . Super .... .. 43’5 • .-s <• 1180 9 10 0 I 1 0 1 2 9 (gain) 30 ■ Super and lime . . 46-5 s ■; 2 O _2 10 2 6 I 14- 0 1 2 3 (gain) 40 Lime. . . 37'2 s 1122 826 0 13 0 0 3 3 '(gain) Control. . 33-4 i 9 1 7 ,6 3

Table 8.

of Paired Plots. Manure. Mean Green Weight of Estimated Weight of Hay per Acre. Value of Hay per ■ Acre. Cost of Manure per Acre. Profit or Loss compared with Unmanured Plot. c rt c. <£ 0 $ Non-sign: or (N). cant lb. T. cwt. qr. £ s. d. £ s. d. £ s. d. 30 Basic slag 14-5 s 0 12 3 3 3 9 O 15 O 013 (loss) 30 Basic slag and lime 13’6 s 0 II 3 2189 180 0 19 3 (loss) 30 Superphosphate . . 16-3 s O 14 I 3 11 . 3 I I O 003 (gain) 30 Super and lime . . i6-8 s 0 14 3 3 13 9 II40 0 10 3 (loss) 40 Lime .'. .. 13-° s on 1 2 16 3 O 13 O 069 (loss) Control. . ' .. . 13-0 n-4 S on 1 0100 2 16 3 2 10 0 O 13 O 0 6 9 (loss)

Table 9.